Culture of intravenous catheter tips has been a routine practice for the past 10 years. Based on the findings of Maki et al. (1977), a quantitative technique based on rolling the tip on a blood agar plate and counting the resulting number of colonies has been a standard practice in most laboratories. The validity and value of these cultures has come into increasing debate, but few well-documented data have been published to clarify this issue. As part of a larger line-maintenance protocol in the Critical Care Medicine Department, our study used these patients to look at colony counts and their predictive value in determining line sepsis from short-term central lines. Approximately 200 patients were enrolled in this study. Organisms from patients who have positive blood cultures and concomitantly positive catheter tip cultures were examined by molecular typing methods to determine if the catheter-tip isolates were identical to the blood-culture isolates, thereby supporting the diagnosis of catheter-related bacteremia. Patient accrual was completed and molecular analysis of the strains of bacteria (strain typing) isolated from these patients was performed. The rate of infection following placement of these lines was low, which resulted in a limited number of episodes available for assessment. Preliminary review of the data suggested that, although patients with positive blood cultures commonly had more than 15 colonies on the rolled catheter culture, there were instances of positive blood cultures when the colony counts were less than 15. In addition, there were cases where the organism cultured from the catheter tip was not the same as that cultured from the blood, and instances where catheter-tip cultures yielded more than 15 colonies that were not associated with positive blood cultures. This project is considered completed.